Compute headline median and eligibility #45

New Issue

2026-09-28T05:42:27Z

xavierk commented

2026-09-28 05:42:27 +00:00

Parent

Odin v1: terminal benchmark and health suite specification

What to build

A complete qualified Standard run can show seven domain scores and a headline median; incomplete or uncalibrated runs show raw values and coverage.

Acceptance criteria

Normalize the approved native workload statistics against immutable mode-specific reference values and aggregate exactly the seven fixed domain memberships.
Take repeated-trial medians in native units first, use the approved even-count rule, and calculate the headline median of seven domain scores.
Withhold a domain/headline for missing required output, detected memory errors, failed verification, safety stop, incomplete required work, VM context, or unqualified calibration; missing health telemetry alone does not suppress a valid score.
Keep loaded responsiveness and health outside numerical aggregation and preserve full score inputs, reference identity, and reasons in output and run record.

Blocked by

Decision sources

## Parent [Odin v1: terminal benchmark and health suite specification](https://git.bongbetic.com/xavierk/odin/issues/26) ## What to build A complete qualified Standard run can show seven domain scores and a headline median; incomplete or uncalibrated runs show raw values and coverage. ## Acceptance criteria - [ ] Normalize the approved native workload statistics against immutable mode-specific reference values and aggregate exactly the seven fixed domain memberships. - [ ] Take repeated-trial medians in native units first, use the approved even-count rule, and calculate the headline median of seven domain scores. - [ ] Withhold a domain/headline for missing required output, detected memory errors, failed verification, safety stop, incomplete required work, VM context, or unqualified calibration; missing health telemetry alone does not suppress a valid score. - [ ] Keep loaded responsiveness and health outside numerical aggregation and preserve full score inputs, reference identity, and reasons in output and run record. ## Blocked by - [Measure CPU throughput](https://git.bongbetic.com/xavierk/odin/issues/31) - [Measure memory bandwidth and loaded response](https://git.bongbetic.com/xavierk/odin/issues/32) - [Check memory errors](https://git.bongbetic.com/xavierk/odin/issues/33) - [Measure GPU compute and rendering](https://git.bongbetic.com/xavierk/odin/issues/35) - [Measure consented storage performance](https://git.bongbetic.com/xavierk/odin/issues/37) - [Measure Bash performance](https://git.bongbetic.com/xavierk/odin/issues/38) - [Measure Rust, C++, and Java text parsing](https://git.bongbetic.com/xavierk/odin/issues/40) - [Complete offline browser suite](https://git.bongbetic.com/xavierk/odin/issues/44) ## Decision sources - [Define Odin’s median score and comparability contract](https://git.bongbetic.com/xavierk/odin/issues/10#issuecomment-6723) - [Define health findings and the optimization advice rubric](https://git.bongbetic.com/xavierk/odin/issues/11#issuecomment-6740) - [Define Odin’s calibration procedure and release gates](https://git.bongbetic.com/xavierk/odin/issues/17#issuecomment-6776) - [Specify Odin’s v1 browser interaction workload](https://git.bongbetic.com/xavierk/odin/issues/20#issuecomment-6927)

xavierk added the ready-for-agent label 2026-09-28 05:42:27 +00:00

xavierk referenced this issue

2026-09-28 05:42:28 +00:00

Browse and compare run history #46

xavierk referenced this issue

2026-09-28 05:42:35 +00:00

Validate eight VM cells #54

xavierk referenced this issue

2026-09-28 05:42:36 +00:00

Collect physical reference calibration #55

xavierk added a new dependency 2026-09-28 05:43:33 +00:00

#31 Measure CPU throughput